I noticed the following behavior yesterday in CB 10.5 Pro (Windows 10) and was wondering, if anyone could help me out:
Whenever I export the Stereo Out that is clipping before the export, let’s say +3.0 dB, I get an exported wave file limited to 0dB - as if there was a limiter added inside the Stereo Out’s Inserts section. The thing is, I don’t have anything on the Stereo Out’s Inserts.
The metering is set to “Post-panner”.
Is there any other setting I need to check in order to get the audio exported without this “automated limiter” feature?
EDIT - this happens on all projects (even new ones).
Thanks a lot.
Hello. In the digital world actually there is nothing above 0 dBFS, you can’t have anything there. In the analog world 0 dB is not the end, you can have more things happening , having some kind of saturation. Actually -18 dBFS digital is the 0 dB analog.
If you work in digital and with floating point though, you could have some signal in the chain (after a plug-in, for example) that goes above 0 dBFS, without clipping, provided this signal doesn’t go to the out this way. That is, internally you can go above 0, but, if you need to go out to your hardware outputs and speakers, the signal must be brought back to be below 0 before this.
thanks for the reply, I understand what you’re saying.
But I think my issue lies with some setting in CuBase - let me clarify it a bit more:
This is what I see before the export:
This is what I get after the export:
Ozone’s there just for metering check purposes - it’s bypassed. And the issue happens without any plugin insterted on the Stereo Out as well.
Well after you Export the Audio, the file produced will be clipped at 0dB (losing the 6.1dB to the clip). So if you import that into a Project it should show 0dB all the way through if the entire signal path is set to unity gain. If I’m not missing something (always a strong possibility) your pics are showing this.
This would be all good, IF there was a limiter inserted on the Stereo Out when this issue occurs - which isn’t the case.
Hence, my Q.
Your description of the “issue” is not clear.
When you say “limiter inserted”, at what point in time?
Do you mean a limiter on the Output at the time of the export? If so, the clipping is acting like a limiter - although a very harsh and unforgiving limiter.
An audio .wav file cannot have a single data point that is above 0dB. So when you try to export any data it will clip anything above 0dB and the resulting file is effectively brick-wall limited to 0dB.
If you mean something different than this, you’ll need a better explanation of what you think is wrong. Because everything you’ve shown so far looks like normal expected behavior.
Try this. Take the file you exported and then imported again into Cubase. Select it and then use Audio>Statistics & post the results here.
I can say too that I see you are trying to mixdown something that is above 0 dBFS and this will be cut at 0, loosing what lays above this, even without limiter on the master. You absolutely need to bring the levels back to bellow 0 before bouncing.
This is bad practice. Get your individual channels to peak no higher than -10dBFS and preserve those peaks throughout the mix chain. By doing this your final mix should end up with a good amount of headroom, which is what you should be aiming for.
thanks for the replies.
yes, that’s what I meant.
Here are the stats (couldn’t upload a .txt file):
This means, that those +6.1 dB which I saw before the export, tells me how much dB will be “clipped/limited” when the export process takes place, right?
My assumtion (obviously a wrong one ) was that the +6.1 dB will be present on the exported file as well - thought I saw it on older CB versions and was afraid that now there’s a built-in limiter, which kicks in if your Stereo Out goes above 0.0 dB and can be switched off somewhere…
Based on Raino’s and Knopf’s advice, I think that the real issue wasn’t in CB, but in my rubbish understanding of the export/metering process.
Well it can be a bit unintuitive. And it doesn’t help that in the analog domain (which was the only game in town for most of the history of recording & source of much conventional wisdom) behaves differently than digital. As an analogy imagine you are stacking boxes in a room with a 7 foot ceiling. With digital you can stack boxes fine until you reach the ceiling and then even the thinnest box can’t be squeezed in. With analog it works the same until you reach the ceiling. Then when you try to squeeze in another package the analog ceiling can stretch a bit and let you fit in more packages. Now those packages at the top are gonna get squished and distorted, but the stack can be taller than 7 feet. Where with digital it’s a hard cutoff at 7 feet.
In both cases once you try and exceed 0dB, distortion will occur. When it is analog distortion it can add some warmth and up to a point sound pleasing as millions of electric guitarists will attest. But digital distortion sounds harsh and ugly - and of course harsh & ugly is exactly what some guitar players want.
This is the best practice. The only place in your signal path that you want to even get close to 0dB (not at or above) is at the output.